6752 ( I 1) Crude yields of chromatographically pure products were uniformly quantitative: the yields reported in this paper are for distilled or recrystallized products. All of the reactions reported herein have been carried out many times with completely reproducible results. (12) A. McKillop, B. P. Swann, and E. C. Taylor, J. Am. Chem. SOC.,95, 3340 (1973). (13) This is true both for 0.1 and 1.0 molar scale reactions.
Edward C. Taylor,* Chih-Shu Chiang Department of Chemistry, Princeton University Princeton, New Jersey 08540 A. McKillop School of Chemical Sciences, University of East Anglia Norwich, England NR4 7TJ J. F. White Emery Industries Cincinnati, Ohio 45232 Received June 1 , 1976
5
0
6a,X=Br b,X=Cl
16
Silver assisted hydrolysis of 1 in aqueous acetone proved more fruitful. In 90% aqueous acetone, 1 reacted eight-ten times slower than 6a, but two-three times faster than 6b. The products isolated from hydrolysis (buffered to allow isolation of 9) are shown in eq 2. YOOH
pOH
93% aqueous > 4
AgC10, (5 equiv) pyridine ( 5 e q u v ) 24 h rcom temp
-
7 72%
Propellanes. 12. A Bridgehead Olefin Transoid in a Six-Membered Ring. Formation of a Stable Cyclopropanone
8
14%
+
+ 9 1.2%
Sir:
A surge of interest' in the production and fate of bridgehead olefins has followed Wiseman's demonstration that these olefins are similar in strain energy and reactivity to their corresponding trans cycloolefins.2 As expected, most work has been limited to cases where the bridgehead double bond is transoid in a ring of 2 7 carbons. While effort has been expended on the synthesis of examples of bridgehead double bonds transoid in six-membered rings,3 no energy comparisons have been possible. We now report data which allow some perspective on the energetics of bridgehead olefin formation. Our approach4 to bridgehead olefins has been through the solvolysis of tricyclic cyclopropyl halides (Le., propellanes); others have also utilized this route.5 An obvious potential precursor for a bridgehead olefin transoid in a six-membered ring is 9,9-dibromo[3.3.1]propellane (9,9-dibromotricyclo[3.3.1.01,5]nonane,1),6 although one would expect relatively more cyclopropyl product than in the earlier cases we reported4b,c(eq 1).
(2)
10 2.1%
Thus the acid mixture of 7 and 8 was the major product. Catalytic hydrogenation (Pd/C, EtOH) gave pure 7.' The olefinic resonance of 8 indicated its presence; the 13C NMR of the mixture showed nine peaks not attributable to 7, including two olefinic carbons (6 151.9, 122.2). The diazomethane-derived esters showed separate resonances at 6 3.62 (7-OCH3) and 6 3.67 (8-OCH3). Cyclopropanone 9, which led to primarily 7 when treated under unbuffered hydrolysis conditions or when shaken with aqueous base, was identified Scheme I
12
Br-Br
1
2 i$OH
Br-OS
Br-YoH
OH
H+
W 4
Solvolysis of 1 (HOAc, 0.012 M NaOAc, 125 "C) proceeded with a rate constant of 3.1 X 10-7s-1, some 6200 times slower than acetolysis of 5. Unfortunately, however, 2 mol of acid was produced, indicating that both bromines had been lost. We pursued this no further, since our experience with 6a indicated we were probably seeing HOAc addition to the cyclopropane ring followed by solvolysis; with 6a, addition took place six times faster than the solvolysis of 1. Journal of the American Chemical Society
/
98:21
Cg attack
-H',
-HBr
- 8
15
/ October 13, 1976
j
~
cleavage
*XH 14
6753
on the basis of its ir (CC14) carbonyl absorption at 1824 cm-I, its mass spectrum (Calcd for C9Hl2O: m/e 136.0888. Found: m / e 136.0883. Calcd for CsH12: m/e 108.0939. Found ( P CO, rel. intensity 1.2): m/e 108.0938. Calcd for C7HgO: m/e 108.0575. Found (P - ClH4, rel. intensity 1.0): m/e 108.0575), and its I3C NMRI2 (CDC13): 6 174.1, 32.8, 30.7, 30.4.I3,l4 Compound loshowed A,, 250 (log €4.23)nm (MI6 250 (log 6 3.95)); also calcd for CgH120: m/e 136.0888. Found: m / e 136.0886. The mode of product formation is shown in Scheme I. Thus ionization of 1 to partially opened ion 1117 is followed by collapse at Cg(kc) to give 4 (S = H), or at C I (CS,kB) to give 3 (S = H). Dehydrobromination of 4 gives 9. While acid formation may funnel through 9, we show the alternative protolysis of the cyclopropane ring of 4. Ion 13 may be derived from 4 either directly or via 14;production of the cis acid (7) is exp e ~ t e d .The ~ ~ less , ~ ~favored formation of unsaturated acid 8 can involve the alternate direction of protolytic cleavage of the Cl -C9 bond with concomitant elimination to aldehyde 15; Tollens oxidation by Ag+ then produces 8.19 Bridgehead olefin 3 suffers a fate similar to those derived from 54bxdand 6.4c While the small amount of 10 formed fits prior expectations for the mode of reaction of 11,the absence of typical bridgehead olefin products 12 and/or 16 was worrisome. Hydrolysis in less aqueous media would be expected to enhance fragmentation to 16;4cindeed Ag+ assisted solvolysis of 1 in 99% aqueous acetone produced roughly equal amounts of 10 and 16 (IH NMR 6 5.80 (s); ir vc-0 at 1690 cm-I. Calcd for C9H130Br:m/e 216.0150. Found: m/e 216.0156). Additionally, the Ag+ assisted acetolysis of 1 gave a derivative of 12 (18),as shown in eq 3. The spectral identification of 17 was made secure via its base catalyzed conversion to 7. Similarly, 18 was converted to 12,and subsequently 10.
lequiv of AgClO, HOAc/Ac,O
OShrmm temp
17
2%
I
OAc 18
2%
In 90% aqueous acetone, the ratio k B / k c , as measured by the percent bridgehead olefin products divided by the percent cyclopropyl products, is 0.024 for 1,1360 for 5, and 1.8 for 6a. The difference between the bridgehead olefins derived from 1 and 5 is that the former is transoid in a six-membered ring, while the latter is transoid in a seven-membered ring; both are cisoid in six-membered rings. The energy required to produce the k e / k c change-roughly 6 kcallmol-is a first approximation to the energy difference between transoid seven and transoid six bridgehead olefins which bear a halogen substituent (the difference for alkyl or hydrogen substituted ones should be greater). Similarly, the difference between analogous cisoid seven and cisoid six bridgehead olefins (5 vs. 6a) is calculated to be ca. 3 kcal/mol.
Dauben and J. D.Robbins, Tetrahedron Lett., 151 (1975); (d) D. Lenoir and J. Firl, Justus Liebigs Ann. Chem., 1467 (1974); (e) W. Burns and M. A. McKeNey, J. Chem. SOC.,Chem. Commun., 858 (1974); (f) A. D. Wolf and M. Jones, Jr., J. Am. Chem. SOC.,95, 8209 (1973); (9) C. B. Quinn, J. R. Wiseman, and J. C. Calabrese, ibid., 95, 6121 (1973); (h) A. H. Aiberts, J. Strating, and H. Wynberg, Tetrahedron Lett., 3047 (1973); (i) J. E. Gano and L. Eizenberg, J. Am. Chem. SOC.,95,972 (1973); (j) H. H. Grootveid, C. Blomberg, and F. Bickelhaupt, J. Chem. SOC.,Chem. Commun., 542 (1973); (k) D. Lenoir, Tetrahedron Left., 4049 (1972); (I) D. Grant, M. A. McKeNey, J. J. Rooney, N. G. Samman. and G. Step, J. Chem. Soc.,Chem. Commun., 1186 (1972); (m) R. Keese and E. P. Krebs, Angew. Chem., lnt. Ed.EngI., 11,518(1972); 10,282(1971);(n)J.O.ReedandW.Lwowski, J. Org. Chem., 36, 2864 (1971). (4) (a) P. Warner, R. LaRose, R. F. Palmer, C. Lee, D. 0. Ross, and J. C. Clardy. J. Am. Chem. SOC., 97, 5507 (1975); (b) P. Warner, S. Lu, E. Myers, P. DeHaven, and R. A. Jacobson. Tetrahedron Lett., 4449 (1975);(c) P. Warner and S.Lu, J. Am. Chem. Soc.. 97, 2536 (1975); (d) P. Warner, J. Fayos. and J. C. Clardy, Tetrahedron Lett., 4473 (1973); (e) P. Warner, R. LaRose, C. Lee, and J. C. Ciardy, J. Am. Chem. SOC.,94, 7607 (1972). (5) (a) C. 8. Reese and M. R. Stables, J. Chem. SOC.,Chem. Commun., 1231 (1972); (b) D. B. Ledlie, T. Swan, J. Pile, and L. Bowers, J. Org. Chem., 41, 419 (1976); (c) D. B. Lediie and L. Bowers, ibid., 40, 792 (1975); (d) D. B. Ledlie, J. Knetzer, and A. Gitterman, ibid., 39, 708 (1974): (e) D. B. Ledlie and J. Knetzer, Tetrahedron Left.. 5021 (1973). (6) P. Warner, R. LaRose, and T. Schieis, Tetrahedron Lett., 1409 (1974). (7) Mp 43-44 OC ( I t 8 40-43 "C). Calcd for C9H1402:m/e 154.0994. Found: m/e 154.0996. The 13C NMR showed only six peaks: 6 188.0, 59.8, 49.8, 38.1, 34.0. 26.3. (8) A. C. Cope and E. S. Graham, J. Am. Chem. SOC.,73,4702 (1951). (9) Compare uc;=o 1822 cm-l for l,ldi-tert-butylcyclopropanonelo and 1825 for trans-1 ,2-di- tert-butylcyciopropanone.' (10) J. K. Crandall and W. H. Machleder, J. Am. Chem. Soc., 90, 7347 (1968). (11) J. F. Pazos, J. G. Pacifici, G. 0. Pierson, D. B. Sclove, and F. D. Greene, J. Org. Chem., 39, 1990 (1974). (12) The carbonyl absorption at 6 174.1 is some 40 ppm upfieid from that of transdi-tert-butyicyclopropanone," which is the only other case reported. In the presence of 2.5 equiv of CrAcAc. the I3C NMR peaks appeared at 6 173.6, 35.3, 33.8, and 29.9. Importantly, we recovered 9 unchanged (ir, 'H NMR) after 13C NMR analysis. (13) Also 'H NMR (CCi4): 6 2.5-1.1 (m); uv (CH&) 325 ( e 27), 336 (c 22) nm. (14) As might be expected from its structure, 9 was relatively inert. It was stable to oxygen and, after stirring for 16 h at room temperature in anhydrous MeOH. was recovered unchanged. After refluxing for 8 h in MeOH, a ca. 75% yield of a very acrid smelling material was obtained. 'H NMR absorptionsat65.35and2.8-1.1,as weilasirpeaksat 1700and 1640cm-', were observed for the unidentified product(s). Furthermore, 9 did not hydrogenate appreciably (i't/C, EtOH, 50 psi, 2 h), although a small peak at 1730 cm-' was observed ( U C = ~ 1726 cm-' for bicyclo[3.3.l]nonan-9one15). (15) C. S.Foote, J. Am. Chem. SOC., 86, 1853 (1964). (16) V. Prelog, K. Schenker, and W. Kung, Helv. Chim. Acta, 36, 471 (1953). (17) (a) U. Scbllkopf, Angew. Chem., int. Ed. Engl., 7, 588 (1968);(b) P. Warner and S.Lu, submitted for publication. (18)An alternate mode for production of 7 and 8 is direct electrophilic attack on 1 by HCIO4. To test this, 1 equiv of EtBr was aiiowed to react with 0.9 equiv of AgCi04 in 90% aqueous acetone. To the resulting 0.9 equiv of HClO4 was added 1 equiv of 1, and the mixture stirred 12 h at room temperature; 92% of 1 was recovered. Even less likely than protonolysis of 1 is Ag' cleavage. When the more susceptible [3.3.l]propeilane was exposed to AgCIO4 in aqueous acetone (which contained 1 equiv of HCi04) for 17 h, only 28% starting material was recovered. However, no other tractable products resulted. We thus feel confident that 1reacts via initial C-Br bond heterolysis. (19) The possibility that 8 arose from 16, i.e.,
'
16
(1) Reviews: (a) R. Keese, Angew. Chem., Int. Ed. Engi., 14, 528 (1975); (b) G. L. Buchanan, Chem. SOC.Rev., 3, 41 (1974); (c) G. Kobrich, Angew. Chem., Int. Ed. Engl., 12, 464 (1973). (2) J. R. Wiseman and W. A. Pletcher, J. Am. Chem. SOC.,92, 956 (1970). 13) (a) A. Nickon. D. F. Covey, F. Huang, and Y.-N. Kuo, J. Am. Chem. SOC., 97,904 (1975); (b) M. Kim and J. D. White, ibid., 97,451 (1975); (c) W. G.
5
-4
15
-
8
OH OH was excluded by a control experiment, as it also was for the analogous nineand ten-membered ring compounds. (20) Fellow of the Alfred P. Sloan Foundation, 1976-1978.
Philip Warner,*zoShih-Lai IA Department of Chemistry, Iowa State Uniuersity Ames, Iowa 5001 I Received June 14, I976
Acknowledgments. We thank the donors of the Petroleum Research Fund, administered by the American Chemical Society, and the Research Corporation for partial support of this work. We are indebted to Mr. Thomas Schleis for the synthesis of generous amounts of 1. References and Notes
($)- @ 'CHBr
Stabilization of Aryldiazonium Ions by Crown Ether Complexation Sir:
Despite an early beginning, the chemistry of aromatic diazonium compounds remains in vogue as additional synthetic Communications t o the Editor
